bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1 Dillenia indica fruit extract has Glucose and Cholesterol Lowering
2 effects
3 Shumsuzzaman Khan1,2,4*, Amrita Bhowmik2, SM Badier Rhaman1, Siew
4 Hua Gan4, Begum Rokeya2
5 1Department of Biochemistry & Molecular Biology, Jahangirnagar University,
6 Savar, Dhaka-1342, Bangladesh.
7 2Department of Pharmacology, Biomedical Research Group, Bangladesh
8 Institute of Research and Rehabilitation in Diabetes, Endocrine and
9 Metabolic Disorders (BIRDEM), Dhaka-1000, Bangladesh.
10 3School of Pharmacy, Monash University Malaysia, Jalan Lagoon Selatan,
11 47500 Bandar Sunway, Selangor, Malaysia.
12
13 4Department of Physiology & Biophysics, School of Medicine, Case Western
14 Reserve University, Ohio, USA.
15
16 *Correspondence: [email protected]; Tel.: +14435384402.
17
18
19
20
21
22
23
24
25
Page 1 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
26 Abstract
27 Background: Dillenia indica (D. indica) can suppress carbohydrates
28 hydrolysis by inhibiting α-amylase and α-glucosidase. However, there is a lack
29 of understanding of its therapeutic potential as an antidiabetic and anti-
30 hyperlipidemic agent.
31 Methods and findings: Type 2 diabetes (T2D) was induced by a single
32 intraperitoneal injection of Streptozotocin (STZ; 90mg/kg) and hyperlipidemia
33 by feeding with 1% cholesterol, 5% coconut oil and 5% cow fat diet.
34 Administration of D. indica extracts in water for four weeks triggered a
35 significant (p≤0.05) reduction in fasting serum glucose (FSG) levels with
36 concomitant improvement in serum insulin levels. Both the water- and
37 ethanol-extract of D. indica treated groups showed significant (p≤0.01)
38 reduction in total cholesterol levels by 25% and 19%, respectively. HDL-
39 cholesterol was also augmented (by 14%) in ethanol-extract treated group.
40 Liver glycogen content was higher in the water-extract treated group.
41 Histopathological examination revealed that there was no tubular epithelial
42 cell degeneration or necrosis in the renal tissues or hepatocyte degeneration
43 and sinusoidal dilation in liver tissues in animals that received the water-
44 extract. On the other hand, consumption of D. indica extract with 1%
45 cholesterol, 5% coconut oil diet or with a 5% cow fat diet for 14 days
46 significantly reduced serum cholesterol levels in group-lll (60→45 mg/dl;
47 p≥0.05) and -IV (85→66 mg/dl; p≥0.05) hypercholesterolemic model rats. D.
48 indica fruit extract also reduced serum TG levels (Group-III: 87→65 mg/dl;
49 Group-IV: 40→90 mg/dl; p≥0.05). Interestingly, treatment with D. indica
Page 2 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
50 prevented a reduction in serum HDL levels in those hypercholesterolemic
51 model rats. Serum LDL levels were significantly lower in group-III (47→39
52 mg/dl; p≥0.05) and group-IV (57→44 mg/dl; p≥0.05) hypercholesterolemic
53 model rats after D. indica treatment.
54 Conclusion: D. indica fruit ameliorates FSG, insulin secretion, glycogen
55 synthesis, and serum lipid profile. Therefore, D. indica fruit can be a potential
56 therapeutic agent for diabetic and hyperlipidemia.
57 Keywords: D. indica fruit; Diabetes; Insulin; Lipid-Profile; Histopathology.
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
Page 3 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
74 Introduction
75 Diabetes mellitus is a chronic metabolic disorder characterized by on-going
76 hyperglycemia. Insulin resistance and decreased insulin secretion are the two
77 cardinal features of type 2 diabetes mellitus (T2DM). T2DM is typically
78 diagnosed based on glucose levels, either a fasting plasma glucose (FPG) ≥
79 7.0 or a 2-hour 75 g oral glucose tolerance test (OGTT) ≥ 11.1 mmol/L levels.
80 Pre-diabetic classifications include a fasting blood glucose (FBG) of 5.6–
81 6.9 mmol/L or a 2-hour blood glucose level of ≥7.8 and < 11.1 mmol/L [1].
82 T2DM is often accompanied by cardiovascular disease, diabetic neuropathy,
83 nephropathy, and retinopathy. An altered lipid profile is common in T2DM
84 patients. Furthermore, reduced hepatic glycogen storage is also observed in
85 diabetes [2-4]. Advancements in the modern lifestyle over the last century
86 have contributed to a dramatic escalation in the incidence of T2DM and
87 hyperlipidemia worldwide [5, 6]. Diabetes mellitus (DM) ranks seventh among
88 the leading causes of death and ranks third when its complications are taken
89 into account [7]. One of the potential complications of T2DM is coronary heart
90 disease due to hyperlipidemia. Hyperlipidemia is a metabolic disorder
91 characterized by successive accumulation of lipids and leukocytes in the
92 arterial wall. This can contribute to many forms of diseases, especially
93 cardiovascular ones such as myocardial infarction and stroke. Moreover,
94 systemic hypercholesterolemia is associated with massive neutrophilia and
95 monocytosis [8-11]. Peripheral leukocyte amount is proportional to the level of
96 cardiovascular complications [12]. Moreover, hypercholesterolemia is
97 positively associated with systemic neutrophil as well as monocyte expansion,
98 suggesting that abnormalities in circulating lipids can influence myeloid cell
Page 4 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
99 expansion [13]. Thus the pathophysiological mechanism underlying
100 hypercholesterolemia is the enrichment and accumulation of systemic
101 neutrophils and monocytes which subsequently increase atherosclerosis.
102 Atherosclerosis due to hyperlipidemia (elevated levels of cholesterol, TG,
103 LDL) is the principal cause of mortality affecting people worldwide [14].
104 According to the World Health Organization (1999), it was estimated that high
105 cholesterol level causes around one-third of all cardiovascular disease
106 worldwide and that there are 10,000,000 people with familial
107 hypercholesterolemia worldwide [15, 16]. Thus, overall prevention and
108 amelioration of T2DM and hyperlipidemia require an integrated, international
109 approach to combat the rapid increase in the number of patients in the
110 forthcoming years [5, 17].
111 Dietary cholesterol has a direct effect on plasma levels of cholesterol [18-20].
112 Dietary cholesterol increases serum cholesterol in all common species if high
113 enough loads are given. The extent of increase depends on the compensatory
114 mechanisms such as enhanced excretion of bile acids and neutral sterols and
115 regulation of cholesterol synthesis and absorption [21]. Lipid-lowering agents
116 have been responsible for a 30% reduction of cardiovascular diseases,
117 therefore validating the search for new therapeutic drugs to reduce
118 hyperlipidemia [22]. In addition, natural products have the potential to prevent
119 T2DM or to keep the disease under control [23-25]. The World Health
120 Organization (WHO) has also recommended evaluating the effectiveness of
121 natural products when there is a lack of safe modern drugs [26, 27]. Moreover,
122 it is believed that natural products may have fewer side effects than
123 conventional drugs. One such natural product is Dillenia indica (D. indica),
Page 5 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
124 locally known as chalta or “elephant apple”. The fruit of D. indica is a 5-12 cm
125 diameter aggregate of 15 carpels, with each carpel containing five seeds
126 embedded in an edible pulp. D. indica fruit is widely used in many indigenous
127 medicinal preparations against several diseases [28] (Supplemental Figure 1).
128 The enzyme inhibition capacity of the active constituents in D. indica has been
129 reported in several studies. For example, betulinic acid in D. indica fruits can
130 inhibit tyrosinase [29], and sterols in D. indica leaves can inhibit α-amylase
131 and α-glucosidase [30]. Nevertheless, despite its traditional claims [31],
132 limited data is available on the hypoglycemic and anti-hyperlipidemic activities
133 exerted by D. indica fruit. Here we hypothesized that D. indica fruit may
134 reduce post-prandial hyperglycemia and hyperlipidemia by suppressing
135 carbohydrate hydrolysis and lipid absorption in the gut respectively, which
136 may be beneficial for diabetic and hyperlipidemic control. Thus, the aim of this
137 present study was to investigate the antidiabetic and anti-hyperlipidemic effect
138 of D. indica fruit in Streptozotocin (STZ)-induced type 2 diabetic model rats
139 and high fat diet (1% cholesterol, 5% coconut oil and 5% cow fat) induced
140 hyperlipidemic model rats.
141 Here we showed that oral administration of both D. indica extract in water or
142 ethanol in Long-Evans female rats for 28 consecutive days significantly
143 ameliorates serum fasting glucose with concomitant enhancement in insulin
144 secretion and liver glycogen content. Moreover, no noticeable change was
145 observed in the serum creatinine and ALT levels in the extract in water and
146 extract in ethanol treated mice, which is reflected in the histopathological
147 analysis, indicating that the extract is safe for kidney and liver functions. In
Page 6 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
148 addition, consumption of D. indica extract with 1% cholesterol and 5%
149 coconut oil or a 5% cow fat diet for 14 days significantly reduced serum
150 cholesterol, TG, LDL levels with concomitant enhancement in HDL in the
151 hyperlipidemic model rats.
152 Methods
153 Plant material
154 Cultivated matured fruits of D. indica (20 kg) were purchased from the local
155 market. Bangladesh National Herbarium ascertained the fruit to be the correct
156 specimen, and a voucher (DACB-35371) was deposited. The fruits were
157 rinsed with fresh sterilized water and sliced into small pieces with a clean
158 knife. D. indica slices were naturally dried under the sun. The dried slices (7
159 kg) were then ground with a blender to yield 1.8 kg of fine powder.
160 Preparation of Extract in Ethanol
161 D. indica powder (900 g) was mixed with 5.4 L of 80% ethanol (1:6), before
162 being kept frozen overnight. On a subsequent day, the suspension was
163 filtered with a sterile cloth, followed by filter paper. Approximately 3.3 L of the
164 filtrate was then collected. The main constituents of D. indica were extracted
165 by using a BUCHI Rotavapor R-114 followed by incubation in a water bath
166 (55˚C) to remove the ethanol. The process yielded approximately 420 ml of
167 filtrate. The semi-dried extract was further dehydrated in a freeze dryer
168 (HETOSICC, Heto Lab Equipment Denmark) at -55˚C followed by storage in
169 an amber bottle (-8˚C). The dried extract was weighed using a digital balance
Page 7 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
170 (GIBERTINI E 42-B). Approximately 150 g (yield: 16.67% w/w) of extract of D.
171 indica fruits was produced.
172 Preparation of Extract in Water
173 The D. indica fine powder (900 g) was dissolved in 9 L water (1:10) and was
174 kept frozen overnight. The suspension was filtered using sterile cloth. The
175 solution was then re-filtered using filter paper to produce approximately 5.5 L
176 of filtrate. The suspension was dried by a BUCHI Rota vapor R-114 and was
177 incubated in a water bath (70˚C) to evaporate the water, yielding
178 approximately 310 mL of filtrate. The semi-dried aqueous extract was further
179 dried in a freeze dryer (HETOSICC, Heto Lab Equipment Denmark) at -55˚C.
180 Subsequently, it was stored in an amber bottle (-8˚C). Ultimately, 90 g of D
181 .indica fruit extract in water was acquired (10% w/w).
182 Animal Model for Anti-Diabetic Study
183 Adult Long-Evans female rats weighing between 170 and 230 g were used.
184 The animals were bred at the Bangladesh Institute of Research and
185 Rehabilitation in Diabetes, Endocrine and Metabolic Disorders (BIRDEM) in
186 the Animal House, Dhaka, Bangladesh. The animals were kept at constant
187 room temperature (22 ± 5˚C), with a humidity of 40-70%, and in a natural 12-
188 hour day-night cycle. The rats were sustained with a standard laboratory
189 pellet diet while water was allowed ad libitum. All animal procedures were
190 performed according to the National Institute of Health (NIH) guidelines under
191 the protocol as approved by the Institutional Animal Care and Use Committee
192 of BIRDEM.
Page 8 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
193 Preparation of Type 2 Diabetes Model Rats
194 STZ (2-deoxy-2-(3-methyl-3-nitrosurea) 1-d-glucopyranose) is a potent
195 alkylating agent that is highly genotoxic and responsible for generating
196 mutations in the cells. STZ contains a highly reactive nitrosurea side chain
197 which is responsible for the initiation of cytotoxic and genotoxic action on
198 pancreatic β-cells. Moreover, STZ is transported into pancreatic β-cells
199 through GLUT2, the glucose transporter, which becomes non-functional in
200 T2DM. Thus, STZ is widely used for the induction of T2DM [32]. Before use, a
201 fresh STZ solution was prepared by dissolving 100 mg STZ in 10 ml (0.1 M)
202 citrate buffer, pH 4-5. The molecular weight for STZ is 265.222, so the
203 molarity for STZ was 3.77 µM. A T2DM condition was created by a single
204 intraperitoneal injection (90 mg/kg) of STZ in rat pups (48 hours old; mean
205 weight 7 g) as previously described [33, 34]. The experiments were done
206 three months after the STZ injection. T2DM can onset at a young age, but
207 most patients are diagnosed at middle age or later. In our study, we focused
208 on adult-onset diabetes by using 3-month-old rats which are an appropriate
209 model for adult-onset. Rats with blood glucose levels of 8–12 mmol/L under
210 fasting conditions were included in the experiments to ensure that STZ had
211 induced type 2 diabetes in all subjects.
212 Grouping of Diabetes Model Rats
213 The rats (n=32) were divided into four groups (n=8). Depending on their
214 group, they were given one treatment per day for 28 consecutive days of one
215 of the following:
Page 9 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
216 1) Type 2 diabetic control group (T2 Control) was given deionized water (10
217 ml/kg).
218 2) Type 2 positive control group (Glibenclamide) was treated with
219 glibenclamide (5 mg/10 ml) (9.9 ml H2O + 0.1 ml Twin 20)/kg.
220 3) Type 2 diabetic extract in water treated group was administered aqueous
221 extract of D. indica (1.25 g/10 ml/kg), which was standardized to 138.89 g
222 fresh fruit of D. indica.
223 4) Type 2 diabetic extract in ethanol treated group was treated with ethanol
224 extract of D. indica at a dose of 1.25 g/10 ml/kg, which was standardized to
225 83.34 g fresh fruit of D. indica.
226 Animal Model for Anti-hyperlipidemic Study
227 The rats (n=24) were divided into four dietary groups (I, II, III and IV) of six
228 rats each. The rats of all groups were fed on a standard pellet diet (diet I) and
229 water ad libitum. Experimental diets were supplied each day through a
230 metallic, smooth stomach tube, along with pellet diet. Rats in group-I were fed
231 with lab pellet diet for 24 consecutive days until the end of the experiment. In
232 order to make the rats hyperlipidemic, the rats from groups-II and -III were fed
233 with 1% cholesterol and 5% coconut oil (diet II) for the first ten days of the
234 experiment. Subsequently, group-II was continued on diet II (1% cholesterol
235 and 5% coconut oil) for an additional 14 days. Rats of group-III were fed with
236 1% cholesterol and 5% coconut oil (for 10 days) in addition to ethanol extract
237 of D. indica for the next 14 days. On the other hand, rats from group-IV
238 received a 5% cow fat diet for the first ten days of experiment and
Page 10 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
239 subsequently were continued with the feeding of 5% cow fat diet along with
240 ethanol extract of D. indica for an additional 14 days.
241 Dose for D. indica Extracts
242 For the oral toxicity study, the Organization for Economic Co-operation and
243 Development (OECD) guideline 425 was followed. Moreover, the
244 histopathological studies on kidney and liver tissues for both extract-treated
245 groups serve as a marker of the safety of D. indica extracts. Treatment with
246 the selected doses for the extract of D. indica in ethanol at 1.25 g/10 ml/kg did
247 not lead to any mortality; all the animals were found to be alive, healthy and
248 active during the experimental periods.
249 Blood Collection Procedure for Biochemical Analysis
250 The animals were anesthetized using isoflurane before blood collection (400
251 µl) following amputation of the tail tip. The samples were centrifuged (2500
252 rpm X 15 minutes), and the serum was transferred to fresh Eppendorf tubes.
253 Serum was refrigerated (-20˚C) until further analysis. All the biochemical
254 experiments were performed within two weeks of serum collection.
255 Collection and Preservation of Liver and Kidney Tissue for Histology
256 Following sacrificed by cervical dislocation, the liver and kidney tissues were
257 collected, washed in normal saline, and fixed by using 40% formaldehyde for
258 24 hours. Subsequently, the tissues were subjected to alcohol dehydration. All
259 tissue samples were washed and embedded by using paraffin and xylene.
260 The tissues were then double-stained.
Page 11 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
261 Determination of serum glucose levels
262 Glucose concentration was estimated by glucose oxidase (GOD-PAP,
263 Boheringer Mannheim GmbH) as described previously [35].
264 Determination of serum insulin levels
265 Serum insulin was analyzed by an enzyme-linked immunosorbent assay
266 (ELISA) kit for rat insulin (Crystal Chem Inc., Downers Grove, IL, USA) [36].
267 Measurement of Liver Glycogen Content
268 The glycogen content in the rat liver was measured by using the anthrone-
269 sulfuric acid method as described previously [37].
270 Determination of serum ALT and creatinine levels
271 Alanine aminotransferase (ALT) levels as well as serum creatinine were
272 measured using a Clinical Biochemistry Analyzer (BioMajesty® JCA-
273 BM6010/C).
274 Haematoxylin & eosin (H & E) staining
275 H & E staining was performed as described previously [38]. In order to
276 demonstrate the difference between the nucleus and the cytoplasm, acid
277 (eosin) and basic (haematoxylin) dyes were used. The slides were placed in
278 Harri’s haematoxylin for 10 minutes and rinsed with tap water until the water
279 was colorless. Then they were counterstained with 1% eosin solution for 1
280 minute. Photomicrographs were acquired by using a transmission microscope
281 (Nikon, Minako, Tokyo) at Dhaka Medical College Hospital, Bangladesh.
Page 12 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
282 Measurement of serum cholesterol levels
283 Serum total cholesterol level was measured by an enzymatic colorimetric
284 method (Cholesterol Oxidase /Peroxidase, CHOD-PAP, Randox Laboratories
285 Ltd., UK), as described previously [39].
286 Determination of serum TG levels
287 Serum triglyceride (TG) was measured by an enzymatic colorimetric glycerol-
288 3-phosphate oxidase phenol aminophenazone (GPO-PAP) method (Randox
289 Laboratories Ltd., UK), as described previously [40].
290 Determination of serum HDL levels
291 Serum HDL level was determined by using an enzymatic colorimetric assay
292 (HDL cholesterol E kit, WAKO Diagnostics), as described previously [41].
293 Determination of serum LDL levels
294 Serum LDL level was calculated either indirectly by using the Friedewald
295 Formula [50] or directly by using the Equal LDL Direct Select Cholesterol
296 Reagent as described previously [41]
297 Statistical Analysis
298 Data were analyzed using a Statistical Package for Social Science software
299 version 12 (SPSS Inc., Chicago, Illinois, USA). The data were reported as
300 mean ± SD or as median (range) where appropriate. Statistical analysis was
301 accomplished by using student t-tests (paired and unpaired) or ANOVA
Page 13 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
302 (analysis of variance) followed by a Bonferroni post hoc test. A p-value of
303 ≤0.05 was considered statistically significant.
304 Results
305 D. indica decreased fasting glucose levels
306 After oral administration of the respective treatments for 28 days, there was a
307 decrease in the fasting serum glucose (FSG) levels of animals in all the
308 groups (Figure 1). Only type 2 diabetic rats treated with extract of D. indica in
309 water showed a significant reduction (p≤0.05) in FSG, although the extract in
310 ethanol treated group showed a reduction of FSG level (by 11%) when
311 compared to the baseline (Day 0). As expected, glibenclamide significantly
312 (p≤0.01) ameliorated the diabetic condition on day 28 by a 23% reduction as
313 compared to the baseline (Day 0). These data suggest that D. indica fruit can
314 lower fasting serum glucose levels.
315 Figure 1. Chronic effect of fruit extracts on fasting serum glucose (FSG)
316 level in STZ-induced type 2 diabetic model rats. Water-extract of D. indica
317 significantly decreased FSG levels in diabetic rats. Results are expressed as
318 mean ± standard deviation (SD). Statistical analysis within groups was
319 conducted using a paired t-test while the comparison between groups was
320 done using a one-way ANOVA with post-hoc Bonferroni correction.*p≤0.05;
321 **p≤0.01.
322 D. indica increased serum insulin levels
323 After 28 days, the extract in water treated group showed a significant (p≤0.01)
324 increase (208%) in serum insulin level, while the type 2 control group showed
Page 14 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
325 a 44% reduction compared to baseline (Day 0). Moreover, the insulin level
326 was significantly (p≤0.05) higher in the extract in water treated group when
327 compared with the type 2 control group (Figure 2). On the other hand, the
328 glibenclamide treated group showed a 30% increase and the extract in
329 ethanol treated group showed a 19% increase in serum insulin levels
330 compared to baseline (Day 0) (Figure 2). These data indicate that D. indica
331 fruit positively modulates pancreatic β-cells to release insulin into the blood.
332 Figure 2. Effect of D. indica fruit extracts in serum insulin levels. Water-
333 extract of D. indica significantly increased serum insulin levels in STZ-induced
334 type 2 diabetic model rats. Results are expressed as mean ± SD. Statistical
335 analysis within the groups was done using a one-way ANOVA with post-hoc
336 Bonferroni correction. *p≤0.05, **p≤0.01.
337 D. indica did not affect body weight
338 The effect of both D. indica extracts (in water and ethanol) on the body weight
339 of type 2 diabetic rats during 28 days of chronic administration was observed.
340 The body weight of each rat was taken at seven days’ intervals and was found
341 to increase on average by 2-7% in all groups. However, there was no
342 significant difference among the groups (Figure 3), suggesting that D. indica
343 fruit extracts did not affect body weight.
344 Figure 3. A consequence of D. indica fruit extracts on rat body weight.
345 No significant change was observed in rat body weight among the groups
346 after chronic treatment with D. indica fruit extracts. Statistical analysis
347 between the groups was done using a one-way ANOVA with post-hoc
348 Bonferroni correction.
Page 15 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
349 D. indica improved liver glycogen content
350 The extract in water treated group showed the highest amount (20.39 mg/g) of
351 liver glycogen content among all the groups, whereas liver glycogen content
352 was lowest (7.54 mg/g) in the type 2 control group. There was no significant
353 difference in the liver glycogen content when the glibenclamide and extract-
354 treated groups were compared (Figure 4). Thus, the fruit of D. indica
355 enhances liver glycogen content in diabetic rats.
356 Figure 4. Effect of D. indica fruit extracts on liver glycogen content. D.
357 indica extract in water showed the highest amount of liver glycogen content
358 among all the groups including the glibenclamide treated group. The results
359 are expressed as mean ± SD.
360 Effect of D. indica on serum lipid profile in diabetic model rats
361 D. indica extract in water significantly decreased (p≤0.01) serum cholesterol
362 on Day 28 [serum cholesterol (mean ± SD) mg/dl: Day 0 (75.00 ± 6.37) vs
363 Day 28 (56.00 ± 4.84)] when compared with the glibenclamide-treated group
364 (Figure 5A). Extract of D. indica in ethanol caused a significant (p≤0.01)
365 reduction in the total cholesterol level on day 28 [serum cholesterol (mean ±
366 SD) mg/dl: Day 0 (75.00 ± 6.30) vs Day 28 (61.00 ± 2.78)] when compared
367 with type 2 control (Figure 5A). Both the extract in water and the extract in
368 ethanol treated groups showed a reduction in serum TG by Day 28, about
369 29% (62.00 to 44.00 mg/dl; p≤0.05) and 32% (57.00 to 39.00 mg/dl; p≤0.05),
370 respectively (Figure 5B).
Page 16 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
371 The extract of D. indica in ethanol significantly increased serum HDL by 14%
372 (36 to 41 mg/dl; p≤0.05), while it reduced LDL by 24% (34 to 26 mg/dl) by Day
373 28 as compared to the baseline (Day 0). In addition, enhanced HDL
374 cholesterol in the extract in ethanol treated group was statistically significant
375 (p≤0.05) when compared with the type 2 control group. Moreover, both the
376 glibenclamide and extract in water treated groups showed an increase in
377 serum HDL levels by 6% (34 to 36 mg/dl) (Figure 5C). Atherogenic LDL-
378 cholesterol levels were decreased by 24% (34 to 26 mg/dl), 19% (36 to 29
379 mg/dl), and 23% (30 to 23 mg/dl) for the extract in ethanol, extract in water,
380 and glibenclamide treated groups, respectively (Figure 5D). These data
381 suggest that D. indica fruit can ameliorate the lipid profile in diabetic rats.
382 Figure 5. Effect of D. indica on serum lipid profile. (A) Both extract in
383 water and in ethanol significantly decreased serum cholesterol levels. (B)
384 Both extracts in water and in ethanol showed a significant reduction in serum
385 TG levels. (C) Extract in ethanol significantly increased serum HDL levels. (D)
386 Reduction in serum LDL levels was also observed by D. indica fruit extracts
387 but was not statistically significant. Results are expressed as mean ± SD.
388 Statistical analysis within the groups was done using a one-way ANOVA with
389 post-hoc Bonferroni correction. *p≤0.05, **p≤0.01.
390 Effects of D. indica on serum ALT and creatinine levels
391 The effects of the extract of D. indica in water or in ethanol on kidney and liver
392 functions were also investigated. There was an increase in the serum
393 creatinine levels only in the type 2 control group, although serum creatinine
394 was decreased in glibenclamide (9%), extract in water (13%), and extract in
Page 17 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
395 ethanol (6%) treated groups (Figure 6A). In addition, there was no significant
396 change in serum ALT levels in any of the groups (Figure 6B). These data
397 indicate that D. indica extracts are safe for kidney and liver functions.
398 Figure 6. Evaluation of D. indica extracts on kidney and liver functions.
399 D. indica fruit extracts led to no significant change in the serum creatinine (A)
400 and the serum ALT (B) levels. Results are expressed as mean ± SD.
401 Statistical analysis within the groups was done using a one-way ANOVA with
402 post-hoc Bonferroni correction.
403 Histopathology
404 To further confirm the safety of D. indica (1.25 g/10 ml/kg) on kidney and liver
405 functions, histopathological examination was performed. Tubular epithelial cell
406 degeneration, necrosis, and hyperemic vessels in the interstitium were
407 examined in the kidney samples. Histological examination revealed that
408 kidney tissues of the type 2 control group were more affected as compared to
409 the extract in water, extract ethanol, and glibenclamide treated groups.
410 Tubular epithelial cell degeneration and tubular epithelial cell necrosis were
411 observed in the kidney tissue of type 2 control group but were absent in the
412 extract in water treated group. The glibenclamide treated group showed well-
413 arranged cells and there was only mild necrosis observed in the extract in
414 ethanol treated group (Figure 7; Table 1), indicating that the aqueous extract
415 of D. indica has some renal protective effects.
416 Figure 7. Histological examination of kidney samples after treatment
417 with D. indica fruit extracts. (A) The normal control group showed well-
418 arranged kidney cells. (B) The type 2 control group (diabetic) showed mild
Page 18 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
419 tubular epithelial cell degeneration and moderate tubular epithelial cell
420 necrosis. (C) The glibenclamide-treated group showed the presence of well-
421 arranged cells. (D) The extract in water treated group showed no toxic effect
422 on kidney cells. (E) The extract in ethanol treated group showed mild
423 necrosis. All figures are observed under a 40x magnification.
424 Table 1: Histopathological changes in kidney samples
Parameters Normal Type 2 Glibenclamide Water Ethanol
group Control group extract extract
group group group
Tubular
epithelial - + - - -
cell
degeneration
Tubular
epithelial - ++ - - +
cell necrosis
Hyperemic
vessels in - - - - -
the
interstitium
425 *Histopathologic assessments of the experimental parameters were graded
426 as follows: (-) showing no change and (+), (++) indicating mild and moderate
427 changes respectively.
428 In the liver samples, hepatocyte degeneration, sinusoidal dilation, and
429 pleomorphism of the hepatocytes were investigated. Histological examination
Page 19 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
430 revealed that the liver tissue of the glibenclamide treated group was more
431 affected as compared to type 2 control, extract in water, and extract in ethanol
432 treated groups. Mild hepatocyte degeneration and sinusoidal dilation were
433 observed in the glibenclamide treated group. Notably, there was no
434 hepatocyte degeneration, sinusoidal dilation, or pleomorphism of the
435 hepatocytes in the extract in water treated group, although mild sinusoidal
436 dilation was observed in the extract in ethanol treated group, suggesting that
437 D. indica aqueous extract is safe for liver function (Figure 8; Table 2).
438 Figure 8. Histological consequences in liver samples after treatment
439 with D. indica fruit extracts. (A) The normal control group showed well-
440 arranged liver cells. (B) The type 2 control group (diabetic) showed no
441 significant changes. (C) The glibenclamide-treated group showed mild
442 hepatocyte degeneration and sinusoidal dilation. (D) The extract in water
443 treated group showed no significant pathological changes in liver cells. (E)
444 The extract in ethanol treated group showed mild sinusoidal dilation. All
445 figures are observed under a 40x magnification.
446 Table 2: Histopathological changes in liver samples
Parameters Normal Type 2 Glibenclam Water Ethanol
group Control ide group extract extract
group group group
Hepatocyte
degenerati - - + - -
on
Page 20 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
Sinusoidal
dilation - - + - +
Pleomorphi
sm of the - - - - -
hepatocyte
.
447 *Histopathologic assessments of the experimental parameters were graded
448 as follows: (-) showing no change and (+), (++) indicating mild and moderate
449 changes respectively.
450 Treatment with D. indica fruit extract significantly reduced serum
451 cholesterol levels in hyperlipidemic model rats
452 Measurement of serum cholesterol is crucial since an altered serum metabolic
453 profile is a potential indicator of many pathological conditions including
454 cardiovascular diseases. In this experiment, we showed that consumption of
455 pellet diet did not enhance serum cholesterol levels in control rats (group I).
456 Moreover, there was also no noticeable change in the control rats among all
457 the groups. In contrast, treatment with 1% cholesterol and 5% coconut oil
458 significantly enhanced serum cholesterol levels at day 10 (41 to 66 mg/dl;
459 p≥0.05), and at day 24 from 66 to 71 mg/dl. Moreover, ANOVA analysis
460 showed that enhancement of serum cholesterol levels in group-ll was
461 significant (p≥0.05) when compared with the normal pellet diet group. These
462 data suggest that consumption of 1% cholesterol and 5% coconut oil has an
463 acute effect on the enhancement of serum cholesterol levels as it was
Page 21 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
464 observed at day 10 and further consumption for another 14 days did not lead
465 to any significant enhancement at day 24 (Figure 9).
466 Figure 9: Effect of D. indica fruit on serum cholesterol levels in
467 hyperlipidemic model rats. After 14 days treatment with D. indica fruit
468 extract in ethanol, significantly reduce serum cholesterol levels in 1%
469 cholesterol and 5% coconut oil diet group and in 5% cow fat treated group.
470 Results are expressed as mean ± standard deviation (SD). Statistical analysis
471 within groups was conducted using a paired t-test while the comparison
472 between the groups was done using a one-way ANOVA with post-hoc
473 Bonferroni correction.*p≤0.05; **p≤0.01.
474 Interestingly, when D. indica fruits extract in ethanol (1.25 mg/kg) was
475 supplemented to the 1% cholesterol and 5% coconut oil treatment in group-lll,
476 those hypercholesterolemic rats showed significant reduction in serum
477 cholesterol levels (60 to 45 mg/dl; p≥0.05), suggesting that D. indica fruit
478 extract has potential to reduce serum cholesterol level (Figure 9).
479 Cow fat (tallow) is primarily made up of triglycerides. In this experiment, a 5%
480 cow fat diet significantly enhanced serum cholesterol levels (38 to 85 mg/dl) in
481 group-lV rats. Moreover, enhancement in serum cholesterol levels was higher
482 (224%) when fed the cow fat diet, than when rats were fed the 1% cholesterol
483 and 5% coconut oil diet (162%), suggesting that cow fat is more potent than
484 1% cholesterol and 5% coconut oil in inducing serum cholesterol levels. As
485 expected, group-IV rats given the cow fat diet with D. indica extract showed a
486 significant reduction (85 to 66 mg/dl; p≥0.05) in cholesterol levels. Although,
487 the reduction was significant, the value (66 mg/dl) was higher than levels (45
Page 22 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
488 mg/dl) in group-III rats (Figure 9). Taken together, these data suggest that
489 extract of D. indica in ethanol possesses an anti-cholesterolemic effect with a
490 saturation capacity depending on the fat content and quality of the diet.
491 Treatment with D. indica fruit extract significantly reduced serum TG
492 levels in hyperlipidemic model rats
493 Like was seen with cholesterol levels, the pellet diet did not enhance TG
494 levels in control rats (Group-I). However, treatment with 1% cholesterol and
495 5% coconut oil for 10 days, enhanced serum TG levels (53 to 80 mg/dl) in
496 group-II rats. Further treatment for an additional 14 days, led to no additional
497 enhancement in serum TG levels, indicating a saturation effect of the extract
498 in lowering TG levels, as was observed with cholesterol levels (Figure 10).
499 Figure 10: Effect of D. indica fruit on serum TG levels in hyper-lipidemic
500 model rats. Consecutive treatment for 14 days with D. indica fruit extract
501 significantly reduced serum TG level in the hypercholesterolemia model rats
502 consumed with 1% cholesterol and 5% coconut oil diet group and in 5% cow
503 fat. Results are expressed as mean ± standard deviation (SD). Statistical
504 analysis within groups was conducted using a paired t-test while the
505 comparison between the groups was done using a one-way ANOVA with
506 post-hoc Bonferroni correction.*p≤0.05.
507 Interestingly, in group-III rats, treatment with 1% cholesterol and 5% coconut
508 oil enhanced serum TG levels at day 10, however, the D. indica extract for an
509 additional 14 days significantly reduced serum TG levels (87 to 65 mg/dl;
510 p≥0.05). Treatment with cow fat diet for 10 days significantly enhanced serum
511 TG levels (40 to 90 mg/dl; p≥0.05) in group-IV. Moreover, ANOVA analysis
Page 23 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
512 showed that enhancement of serum TG levels in group-II and -III by cow fat
513 diet was statistically significant (p≥0.05) compared with group-I rats.
514 Consecutive treatment with D. indica extract for 14 days with the cow fat diet
515 significantly decreased serum TG levels (90 to 66 mg/dl; p≥0.05). Noticeably,
516 the enhancement in serum TG levels by 1% cholesterol, 5% coconut oil and a
517 5% cow fat diet in the group-II and group-III rats was statistically significant
518 (p≥0.05) when compared to group-I rats as revealed by ANOVA analysis
519 (Figure 10). Moreover, reduction in serum TG levels in group-III and group-IV
520 rats by D. indica treatment were similar 75% (87 to 65 mg/dl) and 73.3% (90
521 to 66 mg/dl), suggesting that D. indica extract reduced serum TG levels
522 beyond the fat quality and composition of the diet. Taken together, these data
523 suggest that D. indica fruit extract in ethanol has a strong serum TG lowering
524 effect in the hyper-cholesterolemic rats beyond the quality of the fat diet.
525 Treatment with D. indica fruit extract showed no change in serum HDL
526 levels in hyperlipidemic model rats
527 In this study, the mean serum HDL level was 38 mg/dl in the pellet diet group
528 (Grpup-l), with no significant change throughout the 24 days. In group-II rats,
529 there no significant change in serum HDL levels for 10 days, however,
530 treatment with 1% cholesterol and 5% coconut oil significantly reduced serum
531 HDL levels (33 mg/dl to 25 mg/dl; p≥0.05) at 24 days, suggesting that the 1%
532 cholesterol and 5% coconut oil diet is sufficient to reduce the amount of HDL
533 in a chronic consumption strategy (24 days) (Figure 11). This was unlike
534 pattern seen with the diets in groups-I and -II which were sufficient to enhance
535 cholesterol and TG within 10 days and plateaued by the end of 24 days.
Page 24 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
536 Moreover, ANOVA analysis showed that the reduction of HDL levels in group-
537 II rats was statistically significant (p≥0.05) when compared to group-I (Figure
538 11).
539 Figure 11: Effect of D. indica fruit on serum HDL levels in hyper-
540 lipidemic model rats. D. indica fruit extract maintains normal HDL levels in
541 the hypercholesterolemic model rats. 1% cholesterol and 5% coconut oil diet
542 significantly reduced serum HDL levels. Treatment with D. indica fruit extract
543 prevented reduction in serum HDL levels in group III and IV rats. Results are
544 expressed as mean ± standard deviation (SD). Statistical analysis within
545 groups was conducted using a paired t-test while the comparison between the
546 groups was done using a one-way ANOVA with post-hoc Bonferroni
547 correction.*p≤0.05.
548 Interestingly, in group-III rats, treatment with D. indica fruit extract with the 1%
549 cholesterol and 5% coconut oil diet for 14 consecutive days suppressed a
550 reduction in serum HDL levels (33 mg/dl to 34 mg/dl) as seen in group-II rats.
551 In group-IV rats, the cow fat diet reduced serum HDL levels by 26% (35 mg/dl
552 to 26 mg/dl) at 10 days, and this reduction was higher than the reduction seen
553 in the 1% cholesterol and 5% coconut oil diet in group-III rats (13%; 38 mg/dl
554 to 33 mg/dl). As expected, treatment with D. indica fruit extract with the cow
555 fat diet suppressed the reduction of HDL levels (Figure 11). Taken together,
556 these data suggest that D. indica extract has the potential to maintain normal
557 serum HDL levels in hypercholesterolemic rats.
558 Treatment with D. indica fruit extract significantly reduced serum LDL
559 levels in hyperlipidemic model rats
Page 25 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
560 The relationship between HDL and LDL is inversely proportional; a high level
561 of HDL is beneficial, while a high level of LDL is bad for health. In this
562 experiment, consumption of the pellet diet for 24 days did not lead to any
563 significant change in serum LDL levels in group-I rats. On the contrary,
564 chronic consumption of 1% cholesterol and 5% coconut oil diet for 24 days in
565 group-II rats significantly enhanced serum LDL levels (27 mg/dl to 42 mg/dl;
566 p≥0.05). Although the 1% cholesterol and 5% coconut oil diet for 10 days
567 were not sufficient for significant enhancement of LDL levels, consumption for
568 24 days was sufficient to do so (Figure 12).
569 Figure 12: Effect of D. indica fruit extract on serum LDL levels in
570 hyperlipidaemic model rats. Chronic consumption of D. indica fruit extract
571 for 14 days with 1% cholesterol+5% coconut oil and 5% cow fat diet
572 significantly reduced serum LDL levels in group III and IV rats. Results are
573 expressed as mean ± standard deviation (SD). Statistical analysis within
574 groups was conducted using a paired t-test while the comparison between the
575 groups was done using a one-way ANOVA with post-hoc Bonferroni
576 correction.*p≤0.05.
577
578 Interestingly in group-III rats, chronic consumption of 1% cholesterol and 5%
579 coconut oil with D. indica fruit extract in ethanol for 14 days significantly
580 lowered serum LDL levels (47 mg/dl to 39 mg/dl; p≥0.05), suggesting that D.
581 indica fruit simultaneously reduces cholesterol and LDL levels. As expected,
582 chronic consumption of 5% cow fat diet significantly enhanced serum LDL
583 levels (29 mg/dl to 57 mg/dl, p≥0.05) in the group-IV rats. Moreover, treatment
584 with 5% cow fat diet with D. indica fruit extract significantly suppressed serum
Page 26 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
585 LDL levels (57 mg/dl to 44 mg/dl; p≥0.05) in group-IV rats (Figure 12). Taken
586 together, these data suggest that a 5% cow fat diet is a potent enhancer of
587 serum LDL levels, while D. indica fruit extract ameliorated LDL levels in
588 hypercholesterolemic model rats.
589 Discussion
590 To our knowledge, this is the first study to report on the potential of D. indica
591 extracts in ameliorating diabetes and cholesterol levels. In this study, there
592 was a significant reduction of FSG level in the extract of D. indica in water and
593 glibenclamide treated groups and a non-significant FSG reduction in the
594 extract in ethanol treated group. Our finding confirms that extracts of D. indica
595 in both water and ethanol improved glycemic status in T2DM rats.
596 Alterations in carbohydrates and lipid metabolism are associated with insulin-
597 resistant states, ultimately causing diabetes [42, 43]. In our study, to
598 determine the probable mechanism(s) of the hypoglycemic effect following
599 chronic treatment with D. indica extracts, serum insulin levels of type 2
600 diabetic rats were measured at baseline on Day 0 and compared to levels on
601 Day 28. Four weeks of treatment with D. indica extract in water significantly
602 improved serum insulin levels in type 2 diabetic rats. Thus, it is possible that
603 D. indica aqueous extract may enhance cytoplasmic calcium (Ca2+)
604 responsible for changes in electrical activity in pancreatic β-cells, leading to
605 enhanced insulin secretion. Moreover, it is plausible that D. indica acts on the
606 pancreas to cause a hypoglycemic effect similar to the effects seen with Aloe
607 barbadensis and Litsea glutinosa extracts, both of which ameliorated diabetic
608 conditions [40, 44].
Page 27 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
609 Multiple regulatory mechanisms are involved in glucose homeostasis, and the
610 glucose transporter GLUT2 plays a key role. GLUT2 is expressed in different
611 types of tissues including the liver, kidney, intestine, the pancreatic β-cells,
612 neurons, and astrocytes. Insulin secretion is crucial for GLUT2 activation in
613 pancreatic β-cells, and diabetes causes a significant reduction of GLUT2
614 expression. A known mutation in GLUT2 is responsible for transient juvenile
615 diabetes [45, 46]. Orexin neurons and GABAergic cells in the CNS express
616 GLUT2 [47]. It has been reported that, in a hypoglycemic condition, closure of
617 K+ leak channels and increased activity of AMP-activated protein kinase in
618 GABAergic cells are involved in the activation of GLUT2 to maintain glucose
619 homeostasis [48]. Thus, in contrast, D. indica treatment in hyperglycemic
620 conditions may be responsible for the opening of K+ leak channels and
621 decreased activity of AMP-activated protein kinase, an alternative mechanism
622 to maintain glucose homeostasis in the CNS.
623 It has been reported that D. indica fruits, which are rich in proanthocyanidins,
624 contain a high amount of B-type procyanidins but a lower amount of B-type
625 prodelphinidins [49]. Proanthocyanidin is a class of polyphenols, which are
626 water-soluble in nature, are found mostly in a variety of fruits. In our study, D.
627 indica extract in water conferred stronger glucose-lowering effect than the
628 extract in ethanol. This may be explained by water’s high polarity, leading to
629 an uneven distribution of electron density as compared to ethanol. Moreover,
630 most polyphenols are more soluble in water than in ethanol. However, the
631 extract in ethanol also showed anti-diabetic and significant lipid-lowering
632 activities, so other components present in the extract in ethanol must be
633 playing a role.
Page 28 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
634 At the end of the experimental period, the liver glycogen content was
635 increased in the extract in water treated group. Therefore, it may be
636 hypothesized that the hypoglycemic activity of D. indica in T2DM rats occurs
637 due to the increased uptake of glucose for the formation of glycogen due to
638 enhanced glycogenesis. Additionally, it is also possible that suppression of
639 hydrolysis of carbohydrates by inhibiting α-amylase and α-glucosidase may
640 be another underlying mechanism for the antidiabetic effect of D. indica, as
641 was seen in other studies [50-52]. Glucagon is a peptide hormone secreted
642 from the pancreatic α-cells [53]. It is elevated under stress conditions and
643 helps to increase energy expenditure [54]. The effect of glucagon is opposite
644 that of insulin, i.e. producing an enhanced glucose level in response to
645 hypoglycemia, which may responsible for the development of diabetes [55].
646 Thus, D. indica fruit extracts may decrease glucagon levels, which may be
647 another possible underlying mechanism for reducing serum glucose levels.
648 Additionally, T2DM is associated with a marked imbalance in lipid metabolism
649 [56]. Both the extract in water and the extract in ethanol treated groups had a
650 significant (p≤0.01) decrease in serum cholesterol level. There was also a
651 decrease in serum TG (31%) and LDL (24%) levels, while serum HDL
652 cholesterol (14%) was increased in the extract in ethanol treated group. Thus,
653 extract of D. indica in ethanol has cholesterol-lowering effects in type 2
654 diabetic rats, which is comparative with another study [57].
655 In hyper-cholesterolemic study, consumption of a pellet diet (2.5 Kcal/g) for 24
656 days resulted in no significant changes in serum cholesterol, TG, HDL and
657 LDL levels (Fig. 1-4; 1st panel), consistent with a previous report that said that
658 mice fed with a pellet diet did not become obese [58], while a high fat diet-
Page 29 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
659 induced obesity in rats [59]. Moreover, a hypercaloric pellet diet (6 Kcal/g) led
660 to an increase in body fat, arterial pressure, and high serum glucose, insulin,
661 and leptin, levels, a while normal pellet diet (3.5 Kcal/g) did not lead to an
662 increase in any of these [60]. Thus, we may conclude that the lab pellet diet in
663 this experiment was safe for the rats.
664 A high-fat diet is key step in making a hypercholesterolemic/hyperlipidemic rat
665 [61]. In our experiment, we used a 1% cholesterol and 5% coconut oil, or a 5%
666 cow fat diet to make the rat hypercholesterolemic. It is well established that
667 dietary cholesterol is a potent enhancer of systemic circulating lipids.
668 Moreover, data from population studies reported that dietary cholesterol is
669 atherogenic beyond LDL concentrations in the blood, although other studies
670 reported that high cholesterol consumption causes moderate increases in
671 serum cholesterol levels [62]. In this experiment, we not only added 1%
672 cholesterol but also 5% coconut oil in the diet because coconut oil normally
673 enhances cholesterol and LDL to a greater extent than cholesterol alone, as
674 reported previously [63]. Moreover, coconut oil decreases myocardial capillary
675 density and aggravates cardiomyopathy [64]. Thus 1% cholesterol with 5%
676 coconut oil diet was sufficient to make the rats hypercholesterolemic in this
677 study. Indeed, chronic consumption of a 1% cholesterol with 5% coconut oil
678 diet for 24 days significantly enhanced serum cholesterol levels and serum
679 TG levels in addition to serum LDL levels. Moreover, we also observed a
680 significant reduction in serum HDL levels in rats after 1% cholesterol with 5%
681 coconut oil treatment (Fig. 1-4; 2nd panel). It has been reported that cows fed
682 with tallow showed higher total cholesterol in plasma than cows fed a low-fat
683 diet [65]. Moreover, 21% tallow with 1.25% cholesterol consumption for six
Page 30 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
684 weeks led to renal dysfunction and atherosclerosis in the rats [66]. In addition,
685 a beef tallow diet promoted body fat accumulation and reduced
686 norepinephrine turnover rate in brown adipose tissue by reducing sympathetic
687 activity [67]. Thus, cow fat can be a good atherogenic diet to make rats
688 hypercholesterolemic. In this experiment, a 5% cow fat diet significantly
689 enhanced serum cholesterol and LDL levels after 10 days’ treatment (Fig. 1
690 and 4; 4th panel). Moreover, 5% cow fat also enhanced serum TG levels (Fig
691 2; 4th panel). A further extension of the cow fat effect was a reduction in serum
692 HDL levels (Fig. 4; 4th panel). Taken together, 1% cholesterol with 5%
693 coconut oil or a 5% cow fat diet is enough to make the rat
694 hypercholesterolemic.
695 The results of D. indica on hypercholesterolemic model rats showed that there
696 was a significant decrease in serum cholesterol level in the group treated with
697 1% cholesterol and 5% coconut oil diet and in the group treated with a 5%
698 cow fat diet after 14 days of D. indica extract supplementation (Fig. 1; 3rd and
699 4th panels). Thus D. indica fruit possesses therapeutic potential to reduce
700 serum cholesterol levels. The possible mechanisms behind the reduction in
701 serum cholesterol levels may include proanthocyanidin (plant sterols) which is
702 a class of polyphenols found in a variety of fruits including D. indica [49]. It is
703 well established that plant sterols are a potent therapeutic target for
704 cardiovascular health [68]. In addition, other components present in the D.
705 indica extract besides proanthocyanidins may contribute to the lipid-lowering
706 activity. To maintain lipid homeostasis, cholesterol absorption in the intestine
707 is a significant physiological process [69]. In fact, targeting the cholesterol
708 absorption pathway is one of the most important pharmacological
Page 31 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
709 interventions for hyperlipidemia [70]. A carbohydrate diet is positively
710 correlated with serum lipid levels [71]. Chromatographic separation of D.
711 indica leaves extract revealed the existence of betulinic acid, quercetin, β
712 sitosterol, and stigmasterol palmitate [72], all of which significantly inhibited
713 the activities of α-amylase and α-glucosidase, which may contribute to the
714 lipid-lowering effects. Phytosterols (Proanthocyanidins) can cross the blood-
715 brain barrier (BBB) [73]. Therefore, it may possible that proanthocyanidins in
716 D. indica inhibit HMG-CoA reductase and reduce cholesterol biosynthesis. It
717 has been reported that bioactive palmitoleate can prevent atherosclerosis by
718 suppressing organelle stress and inflammasome activation [74]. Moreover, it
719 has also been reported that phytosterols in D. indica possess antioxidant [49]
720 and anti-inflammatory activities by inhibiting the production of tumor necrosis
721 factor-alpha [75] and may be the crucial underlying mechanism to prevent
722 atherosclerosis. Microscopy and phytochemical analysis indicated the
723 presence of xylem fibers, phytosterols, pro-anthocyanidins, terpenoids,
724 glycosides, fatty acids, flavonoids, and phenolic compounds in D. indica [49,
725 76]. D. indica is also rich in different types of soluble and insoluble dietary
726 fibers. Phytosterols reduce cholesterol levels by different mechanisms, such
727 as by competing with cholesterol absorption in the gut and in dietary mixed
728 micelles. Additionally, sterols compete with cholesterol for solubilization, co-
729 crystallize with cholesterol to form insoluble mixed crystals and inhibit
730 cholesterol hydrolysis by lipases [77-79]. Therefore, phytosterols in D. indica
731 extracts may inhibit cholesterol absorption and hydrolysis in the gut.
732 Hypercholesterolemic model rats treated with D. indica extract for 14 days
733 showed a significant reduction in serum TG levels in the 1% cholesterol and
Page 32 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
734 5% coconut oil or the 5% cow fat diet group (Fig. 2; 3rd and 4th panels). D.
735 indica extract may interfere with TG biosynthesis and absorption. Overall,
736 lipase is a critical enzyme to digest dietary TGs. Thus the sterols in D. indica
737 fruit extract in ethanol may inhibit lipase, thereby ameliorating fat
738 malabsorption in the intestinal lumen by forming an irreversible bond with fat
739 molecules and ultimately leading to fecal excretion without degradation.
740 Moreover, it has been reported that polyphenols in boysenberry and okra
741 extract suppressed enhancement of TG in plasma [80, 81]; this is consistent
742 with our study, where polyphenol in D. indica fruit may lower TG levels in
743 hypercholesterolemic rats. Diacylglycerol acyltransferase 2 (DGAT2)
744 catalyzes the last step of TG biosynthesis in rodents [82]. Thus, we do not
745 exclude the possibility that the D. indica extract may inhibit DGAT2 and
746 reduce hepatic TG synthesis.
747 Chronic consumption of a 1% cholesterol and 5% coconut oil diet or a 5% cow
748 fat diet for 24 days significantly reduced serum HDL levels but treatment with
749 D. indica fruit extract suppressed reduction and restored normal HDL levels
750 (Fig. 3). ATP-binding cassette protein A1 (ABCA1), lecithin-cholesterol
751 acyltransferase, and apolipoprotein (apo) A-I are the key components for HDL
752 biosynthesis [83]. ABCA1 exports cholesterol from the membranes to the
753 nascent HDL, while ABCG1 (G Member sub-family of ATP-binding Cassette
754 transporter) transports cholesterol to mature HDL; the key features of
755 cholesterol homeostasis [84]. One possible mechanism to reduce HDL levels
756 with chronic consumption of 1% cholesterol and 5% coconut oil is by
757 downregulation of the genes associated with ABCA1, LCAT, and APOAI
Page 33 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
758 molecules. On the contrary, D. indica fruit extract could restore and/or up-
759 regulate these gene functions to maintain HDL homeostasis
760 Both a 1% cholesterol and 5% coconut oil diet, and a 5% cow fat diet
761 significantly enhanced serum LDL levels, and, more importantly, D. indica fruit
762 extract significantly reduced serum LDL levels in both groups (Fig. 4). To
763 investigate the underlying mechanism of such reduction, it is important to
764 understand the pathophysiology of LDL. After entering into the artery wall
765 endothelium, LDL initiates atherosclerotic plaque formation by binding to
766 glycosaminoglycans [85]. LDL is oxidized, thus producing modified apoB by
767 altering lysine residues [86]. The modified LDL is recognized by the
768 macrophages and internalized by endocytosis, forming foam cells [87]. These
769 foam cells release cytokines, initiating inflammatory reactions [88]. As a
770 consequence, the cells of the arterial wall proliferate and produce collagen.
771 The plaque enlarges, eventually leading to blood clot formation and blockage
772 of the vessel. Several studies have reported that HDL prevents LDL oxidation,
773 a crucial initial step in LDL pathophysiology, and, thus, reducing LDL levels
774 [89]. In this experiment, D. indica fruit extract maintained normal HDL levels in
775 hypercholesterolemic model rats, which may contribute to reducing LDL levels
776 by inhibiting LDL oxidation. In addition, LDL enhances CXCR2 (cytokine)
777 expression, facilitating neutrophils accumulation [90]. On the other hand,
778 monocytes use CCR2 for entry into the atherogenic lesion [91, 92]. Therefore,
779 it may possible that D. indica fruit extract may reduce CXCR2 and CCR2
780 expression and reduce atherosclerosis.
Page 34 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
781 The muscles use creatine to make energy and produce creatinine as a waste
782 product. High levels of creatinine in the blood may indicate diabetic
783 nephropathy. Hyperglycemia enhances the level of reactive oxygen species,
784 which then facilitate the formation of glycation end products as implicated in
785 the pathogenesis of diabetic nephropathy [93, 94]. In our study, treatment with
786 D. indica extracts suppressed enhancement of creatinine levels in diabetic
787 rats. On the other hand, ALT is an indicator of liver function. It has been
788 reported that elevated serum levels of ALT are significantly associated with
789 increased diabetic risk [95, 96]. In this study, there was no significant change
790 in serum ALT levels in the D. indica extract-treated groups, indicating that D.
791 indica does not affect liver function. These observations are correlated with
792 the histological changes as extract of D. indica in water showed no tubular
793 epithelial cell degeneration, necrosis, and hyperemic vessels in the
794 interstitium of the kidney tissues and no hepatocyte degeneration, sinusoidal
795 dilation, and pleomorphism of the hepatocytes in liver tissues. Interestingly,
796 the glibenclamide-treated group showed mild hepatocyte degeneration and
797 sinusoidal dilation in our study. Moreover, it has been reported that
798 glibenclamide reduced pro-inflammatory cytokine production and reduced
799 glutathione levels in diabetic patients [97], which be the underlying
800 mechanism of observed hepatocyte degeneration and sinusoidal dilation in
801 our study. Thus, our study not only defines an emerging role of D. indica fruit
802 as an anti-diabetic and anti-hyperlipidemic agent but also opens a new
803 window on the safety concerns of glibenclamide treatment.
804 Conclusions
Page 35 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
805 D. indica fruit has a glucose-lowering effect and enhances insulin secretion as
806 well as glycogen synthesis. It decreases total cholesterol levels and increases
807 HDL-cholesterol. It does not affect renal and liver functions in terms of
808 creatinine and ALT, and, therefore, has the potential to be an anti-diabetic and
809 cholesterol-lowering agent.
810
811
812
813
814
815
816
817
818
819
820
821
822
823
Page 36 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
824 Acknowledgments: The authors are grateful to the technical staff of the
825 Research Division of Bangladesh Institute of Research and Rehabilitation in
826 Diabetes, Endocrine and Metabolic Disorder (BIRDEM) for their assistance.
827 The authors offer thanks to Dr. Zillur Rhaman Khan and Dr. Hemale from
828 Dhaka Medical College for histopathology. The authors are also grateful to
829 Mr. Omrito and Dr. Karim from the Bangladesh Rice Research Institute for
830 their contribution to grinding the sun-dried D. indica.
831 Funding: This work was supported by the International Science Programme
832 (ISP) at Uppsala University, Sweden, through Grant No ISP2016/26:8 to the
833 Asian Network of Research on Antidiabetic Plants (ANRAP).
834 Author Contributions: S.K and B.R designed the research. S.K performed
835 all the experiments and carried out all the data analysis and figure preparation
836 except ELISA experiment by B.R in Fig. 3. A.B contributed to data analysis
837 (SPSS) and the collection of biological samples (blood, liver, and kidney). S.K;
838 S.H.G; SM.B.R; and B.R wrote and edited the manuscript.
839 Edited by: Renee Cockerham ([email protected]), Program
840 Manager, Office of Postdoctoral Scholars, University of Maryland, Baltimore,
841 USA.
842 Conflicts of Interest: The authors declare no conflict of interest.
843
844
845
846
Page 37 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
847 References
848 1. American Diabetes A. 2. Classification and diagnosis of diabetes:
849 standards of medical care in diabetes-2019. Diabetes care 2019; 42:S13-S28.
850 2. Lyssenko V, Almgren P, Anevski D, Perfekt R, Lahti K, Nissén M.
851 Predictors of and longitudinal changes in insulin sensitivity and secretion
852 preceding onset of type 2 diabetes. Diabetes 2005; 54(1): 166-74.
853 3. Wu Y, Ding Y, Tanaka Y, Zhang W. Risk factors contributing to type 2
854 diabetes and recent advances in the treatment and prevention. Int J Med Sci.
855 2014; 11(11): 1185-200.
856 4. Ferrannini E, Lanfranchi A, Rohner JF, Manfredini G, Van-de WG.
857 Influence of long-term diabetes on liver glycogen metabolism in the rat.
858 Metabolism 1990; 39(10): 1082-8.
859 5. Zimmet P, Alberti KG, Shaw J. Global and societal implications of the
860 diabetes epidemic. Nature 2001; 414:782-787.
861 6. Franks PW, McCarthy MI. Exposing the exposures responsible for type 2
862 diabetes and obesity. Science 2016; 354:69-73.
863 7. Marathe CS, Rayner CK, Bound M, Checklin H, Standfield S, Wishart J,
864 Lange K, Jones KL, Horowitz M. Small intestinal glucose exposure
865 determines the magnitude of the incretin effect in health and type 2 diabetes.
866 Diabetes 2014; 63:2668-2675.
867 8. Soehnlein O, Swirski FK. Hypercholesterolemia links hematopoiesis with
868 atherosclerosis. Trends Endocrinol Metab. 2013; 24:129-136.
869 9. Rader DJ, Daugherty A. Translating molecular discoveries into new
870 therapies for atherosclerosis. Nature. 2008; 451:904-913.
Page 38 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
871 10. Lin J, Yang R, Tarr PT, Wu PH, Handschin C, Li S, et al. Hyperlipidemic
872 Effects of Dietary Saturated Fats Mediated through PGC-1β Coactivation of
873 SREBP. Cell. 2005;120:261-273.
874 11. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease.
875 N Engl J Med. 2005; 352:1685-1695.
876 12. Friedman GD, Klatsky AL, Siegelaub AB. The leukocyte count as a
877 predictor of myocardial infarction. N Engl J Med. 1974; 290:1275-1278.
878 13. Rothe G, Gabriel H, Kovacs E, Klucken J, Stohr J, Kindermann W, et al.
879 Peripheral blood mononuclear phagocyte subpopulations as cellular markers
880 in hypercholesterolemia. Arterioscler Thromb Vasc Biol. 1996; 16:1437-1447.
881 14. Lusis AJ. Atherosclerosis. Nature 2000; 407:233-241.
882 15. Garcia CK. Autosomal Recessive Hypercholesterolemia Caused by
883 Mutations in a Putative LDL Receptor Adaptor Protein. Science 2001;
884 292:1394-1398.
885 16. Abifadel M, Varret M, Rabès JP, Allard D, Ouguerram K, Devillers M, et al.
886 Mutations in PCSK9 cause autosomal dominant hypercholesterolemia. Nat
887 Genet. 2003; 34:154-156.
888 17. Bohula EA, Scirica BM, Inzucchi SE, McGuire DK, Keech AC, Smith SR,
889 Kanevsky E, Murphy SA, Leiter LA, Dwyer JP, et al. Effect of lorcaserin on
890 prevention and remission of type 2 diabetes in overweight and obese patients
891 (CAMELLIA-TIMI 61): a randomised, placebo-controlled trial. Lancet 2018;
892 392:2269-2279.
893 18. Djousse L, Gaziano JM. Egg consumption and risk of heart failure in the
894 Physicians' Health Study. Circulation 2008;117:512-516.
Page 39 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
895 19. Ohlsson L, Burling H, Nilsson A. Long term effects on human plasma
896 lipoproteins of a formulation enriched in butter milk polar lipid. Lipids Health
897 Dis. 2009; 8:44.
898 20. Jain MK, Rogers J, Berg O, Gelb MH. Interfacial catalysis by
899 phospholipase A2: activation by substrate replenishment. Biochemistry. 1991;
900 30:7340-7348.
901 21. Dietschy JM, Wilson JD. Regulation of cholesterol metabolism. 3. The N
902 Engl J Med. 1970; 282:1241-1249.
903 22. Nissen SE, Tuzcu EM, Schoenhagen P, Crowe T, Sasiela WJ, Tsai J, et
904 al. Reversal of Atherosclerosis with Aggressive Lipid Lowering I: Statin
905 therapy, LDL cholesterol, C-reactive protein, and coronary artery disease. N
906 Engl J Med. 2005; 352:29-38.
907 23. Bailey CJ, Day C. Traditional plant medicines as treatments for diabetes.
908 Diabetes care 1989; 12:553-564.
909 24. Gushiken LF, Beserra FP, Rozza AL, Bérgamo PL, Bérgamo DA,
910 Pellizzon CH. Chemical and biological aspects of extracts from medicinal
911 plants with antidiabetic effects. Rev Diabet Stud. 2016; 13:96-112.
912 25. Tag H, Kalita P, Dwivedi P, Das AK, Namsa ND. Herbal medicines used in
913 the treatment of diabetes mellitus in Arunachal Himalaya, northeast, India. J
914 Ethnopharmacol. 2012; 141:786-795.
915 26. Bhamra SK, Slater A, Howard C, Johnson M, Heinrich M. The use of
916 traditional herbal medicines amongst South Asian diasporic communities in
917 the UK. Phytother Res. 2017; 31:1786-1794.
918 27. Qian F, Korat AA, Malik V, Hu FB. Metabolic effects of monounsaturated
919 fatty acid-enriched diets compared with carbohydrate or polyunsaturated fatty
Page 40 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
920 acid-enriched diets in patients with type 2 diabetes: a systematic review and
921 meta-analysis of randomized controlled trials. Diabetes care 2016; 39:1448-
922 1457.
923 28. Biswas R, Chanda J, Kar A, Mukherjee PK. Tyrosinase inhibitory
924 mechanism of betulinic acid from Dillenia indica. Food Chem. 2017; 232:689-
925 696.
926 29. Biswas R, Chanda J, Kar A, Mukherjee PK. Tyrosinase inhibitory
927 mechanism of betulinic acid from Dillenia indica. Food Chem. 2017; 232:689-
928 696.
929 30. Kumar S, Kumar V, Prakash O. Enzymes inhibition and antidiabetic effect
930 of isolated constituents from Dillenia indica. Biomed Res Int. 2013;
931 2013:382063.
932 31. Saiful YL, Armania N. Dillenia species: A review of the traditional uses,
933 active constituents and pharmacological properties from pre-clinical studies.
934 Pharm Biol. 2014; 52:890-897.
935 32. Bolzan AD, Bianchi MS. Genotoxicity of streptozotocin. Mutat Res. 2002;
936 512:121-134.
937 33. Bonner-Weir S, Trent DF, Honey RN, Weir GC. Responses of neonatal rat
938 islets to streptozotocin: limited B-cell regeneration and hyperglycemia.
939 Diabetes 1981; 30:64-69.
940 34. Ariza L, Zaguirre M, García M, Blasco E, Rabanal RM, Bosch A, Otaegui
941 PJ. Hyperglycemia and hepatic tumors in ICR mice neonatally injected with
942 streptozotocin. Lab Anim 2014; 43:242-249.
943 35. Basit A, Hydrie MZ, Ahmed K, Hakeem R. Prevalence of diabetes,
944 impaired fasting glucose and associated risk factors in a rural area of
Page 41 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
945 Baluchistan province according to new ADA criteria. J Pak Med Assoc. 2002;
946 52(8): 357-60.
947 36. Kratzsch J, Ackermann W, Keilacker H, Besch W, Keller E. A sensitive
948 sandwich enzyme immunoassay for measurement of insulin on microtiter
949 plates. Exp Clin Endocrinol. 1990; 95(2):229-36.
950 37. Van DVJ. Two methods for the determination of glycogen in liver.
951 Biochem J. 1954; 57(3):410–416.
952 38. Feldman AT, Wolfe D. Tissue processing and hematoxylin and eosin
953 staining. Methods Mol Biol. 2014; 1180:31-43.
954 39. Salmabadi Z, Mohseni KH, Parivar K, Karimzadeh L. Effect of Grape Seed
955 Extract on Lipid Profile and Expression of Interleukin-6 in Polycystic Ovarian
956 Syndrome Wistar Rat Model. Int J Fertil Steril. 2017; 11:176-183.
957 40. Moniruzzaman M, Rokeya B, Ahmed S, Bhowmik A, Khalil MI, Gan SH. In
958 vitro antioxidant effects of Aloe barbadensis Miller extracts and the potential
959 role of these extracts as antidiabetic and antilipidemic agents on
960 streptozotocin-induced type 2 diabetic model rats. Molecules. 2012;
961 17:12851-12867.
962 41. Tadin SM, Peterson LB, Cumiskey AM, Rosa RL, Mendoza VH, Castro
963 PJ, et al. siRNA-induced liver ApoB knockdown lowers serum LDL-cholesterol
964 in a mouse model with human-like serum lipids. J Lipid Res. 2011; 52:1084-
965 1097.
966 42. Rorsman P, Ashcroft FM. Pancreatic beta-cell electrical activity and insulin
967 secretion: of mice and men. Physiol Rev. 2018, 98:117-214.
968 43. Yang Q, Vijayakumar A, Kahn BB. Metabolites as regulators of insulin
969 sensitivity and metabolism. Nat Rev Mol Cell Biol. 2018, 19:654-672.
Page 42 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
970 44. Zhang X, Jin Y, Wu Y, Zhang C, Jin D, Zheng Q, Li Y. Anti-hyperglycemic
971 and anti-hyperlipidemia effects of the alkaloid-rich extract from barks of Litsea
972 glutinosa in ob/ob mice. Sci Rep. 2018; 8(1):12646.
973 45. Thorens B. GLUT2, glucose sensing and glucose homeostasis.
974 Diabetologia 2014; 58:221-232.
975 46. Thorens B, Weir GC, Leahy JL, Lodish HF, Bonner WS. Reduced
976 expression of the liver/beta-cell glucose transporter isoform in glucose-
977 insensitive pancreatic beta cells of diabetic rats. Proc Natl Acad Sci USA
978 1990; 87: 6492-6496.
979 47. Gonzalez JA, Jensen LT, Fugger L, Burdakov D. Metabolism-Independent
980 Sugar Sensing in Central Orexin Neurons. Diabetes 2008; 57:2569-2576.
981 48. Lamy CM, Sanno H, Labouèbe G, Picard A, Magnan C, Chatton JY,
982 Thorens B. Hypoglycemia-Activated GLUT2 Neurons of the Nucleus Tractus
983 Solitarius Stimulate Vagal Activity and Glucagon Secretion. Cell Metab 2014;
984 19:527-538.
985 49. Fu C, Yang D, Peh WY, Lai S, Feng X, Yang H. Structure and antioxidant
986 activities of proanthocyanidins from elephant apple (Dillenia indica Linn.). J
987 Food Sci. 2015; 80:2191-2199.
988 50. Olennikov DN, Chirikova NK, Kashchenko NI, Nikolaev VM, Kim SW,
989 Vennos C. Bioactive phenolics of the genus Artemisia (Asteraceae): HPLC-
990 DAD-ESI-TQ-MS/MS Profile of the Siberian species and their inhibitory
991 potential against alpha-amylase and alpha-glucosidase. Front Pharmacol.
992 2018; 9:756.
Page 43 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
993 51. Khan MF, Rawat AK, Khatoon S, Hussain MK, Mishra A, Negi DS. In vitro
994 and in vivo antidiabetic effect of extracts of Melia azedarach, Zanthoxylum
995 alatum, and Tanacetum nubigenum. Integr Med Res. 2018; 7(2):176-183.
996 52. Ota A, Ulrih NP. An overview of herbal products and secondary
997 metabolites used for management of type two diabetes. Front Pharmacol.
998 2017; 84:36.
999 53. Chang SK, Kohlgruber AC, Mizoguchi F, Michelet X, Wolf BJ, Wei K, Lee
1000 PY, Lynch L, Duquette D, Ceperuelo MV, et al. Stromal cell cadherin-11
1001 regulates adipose tissue inflammation and diabetes. J Clin Invest 2017;
1002 127(9): 3300-3312.
1003 54. Jones BJ, Tan T, Bloom SR. Minireview: Glucagon in stress and energy
1004 homeostasis. Endocrinology 2012; 153(3):1049-54.
1005 55. Chen Q, Mo R, Wu N, Zou X, Shi C, Gong J, Li J, Fang K, Wang D, Yang
1006 D, et al. Berberine ameliorates diabetes-associated cognitive decline through
1007 modulation of aberrant inflammation response and insulin signaling pathway
1008 in DM rats. Front Pharmacol. 2017; 8:334.
1009 56. Gadi R, Samaha FF. Dyslipidemia in type 2 diabetes mellitus. Curr Diab
1010 Rep. 2007; 7(3):228-234.
1011 57. Xie ZQ, Liang G, Zhang L, Wang Q, Qu Y, Gao Y, Lin LB, Ye S, Zhang J,
1012 Wang H, et al. Molecular mechanisms underlying the cholesterol-lowering
1013 effect of Ginkgo biloba extract in hepatocytes: a comparative study with
1014 lovastatin. Acta Pharmacol Sin. 2009; 30:1262-1275.
1015 58. Desmarchelier C, Ludwig T, Scheundel R, Rink N, Bader BL, Klingenspor
1016 M, et al. Diet-induced obesity in ad libitum-fed mice: food texture overrides the
1017 effect of macronutrient composition. Br J Nutr. 2013;109:1518-1527.
Page 44 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1018 59. Woods SC, Seeley RJ, Rushing PA, D'Alessio D, Tso P. A controlled high-
1019 fat diet induces an obese syndrome in rats. J Nutr. 2003;133:1081-1087.
1020 60. Nascimento AF, Sugizaki MM, Leopoldo AS, Lima LAP, Luvizotto RA,
1021 Nogueira CR, et al. A hypercaloric pellet-diet cycle induces obesity and co-
1022 morbidities in Wistar rats. Arq Bras Endocrinol Metabol. 2008; 52:968-974.
1023 61. Matheus VA, Monteiro L, Oliveira RB, Maschio DA, Collares BCB,
1024 Butyrate reduces high-fat diet-induced metabolic alterations, hepatic steatosis
1025 and pancreatic beta cell and intestinal barrier dysfunctions in prediabetic
1026 mice. Exp Biol Med. 2017; 242:1214-1226.
1027 62. Grundy SM. Does Dietary Cholesterol Matter? Curr Atheroscler Rep 2016;
1028 18:68.
1029 63. Bunzel B. Psychological problems and management of cancer patients
1030 from the viewpoint of the clinical psychologist. Wien Med Wochenschr. 1989;
1031 139:498-500.
1032 64. Muthuramu I, Amin R, Postnov A, Mishra M, Jacobs F, Gheysens O, et al.
1033 Coconut Oil Aggravates Pressure Overload-Induced Cardiomyopathy without
1034 Inducing Obesity, Systemic Insulin Resistance, or Cardiac Steatosis. Int J Mol
1035 Sci. 2017; 18.
1036 65. Son J, Grant RJ, Larson LL. Effects of tallow and escape protein on
1037 lactational and reproductive performance of dairy cows. J Dairy Sci. 1996;
1038 79:822-830.
1039 66. Rashid KM, Ahsan H, Siddiqui S, Siddiqui WA. Tocotrienols have a
1040 nephroprotective action against lipid-induced chronic renal dysfunction in rats.
1041 Ren Fail. 2015; 37:136-143.
Page 45 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1042 67. Matsuo T, Shimomura Y, Saitoh S, Tokuyama K, Takeuchi H, Suzuki M.
1043 Sympathetic activity is lower in rats fed a beef tallow diet than in rats fed a
1044 safflower oil diet. Metabolism. 1995; 44:934-939.
1045 68. Pascual FV. Usefulness of plant sterols in the treatment of
1046 hypercholesterolemia. Nutr Hosp. 2017; 34:62-67.
1047 [69] Li Y, Qin G, Liu J, Mao L, Zhang Z, Shang J. Adipose tissue regulates
1048 hepatic cholesterol metabolism via adiponectin. Life Sci. 2014; 118(1):27-33.
1049 70. Ros E. Intestinal absorption of triglyceride and cholesterol. Dietary and
1050 pharmacological inhibition to reduce cardiovascular risk. Atherosclerosis
1051 2000; 151(2):357-379.
1052 71. Ma Y, Li Y, Chiriboga DE, Olendzki BC, Hebert JR, Li W, Leung K, Hafner
1053 AR, Ockene IS. Association between carbohydrate intake and serum lipids. J
1054 Am Coll Nutr. 2006; 25(2):155-163.
1055 72. Kumar S, Kumar V, Prakash OM. Microscopic evaluation and
1056 physiochemical analysis of Dillenia indica leaf. Asian Pac J Trop Biomed.
1057 2011; 1(5):337-340.
1058 73. Burg VK, Grimm HS, Rothhaar TL, Grosgen S, Hundsdorfer B,
1059 Haupenthal VJ, et al. Plant sterols the better cholesterol in Alzheimer's
1060 disease? A mechanistical study. J Neurosci. 2013; 33:16072-16087.
1061 74. Çimen I, Kocatürk B, Koyuncu S, Tufanlı Ö, Onat UI, Yıldırım AD, et al.
1062 Prevention of atherosclerosis by bioactive palmitoleate through suppression of
1063 organelle stress and inflammasome activation. Sci Transl Med. 2016;
1064 8:358ra126-358ra126.
Page 46 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1065 75. Somani SJ, Badgujar LB, Sutariya BK, Saraf MN. Protective effect of
1066 Dillenia indica L. on acetic acid induced colitis in mice. Indian J Exp Biol.
1067 2014; 52:876-881.
1068 76. Nguyen TT. The cholesterol-lowering action of plant stanol esters. J Nutr
1069 1999; 129(12):2109-2112.
1070 77. De SE, Mensink RP, Plat J. Effects of plant sterols and stanols on
1071 intestinal cholesterol metabolism: suggested mechanisms from past to
1072 present. Mol Nutr Food Res. 2012; 56(7):1058-1072.
1073 78. Hunter PM, Hegele RA. Functional foods and dietary supplements for the
1074 management of dyslipidaemia. Nat Rev Endocrinol. 2017; 13(5):278-288.
1075 79. Trautwein EA, Koppenol WP, de JA, Hiemstra H, Vermeer MA, Noakes M,
1076 Luscombe MND. Plant sterols lower LDL-cholesterol and triglycerides in
1077 dyslipidemic individuals with or at risk of developing type 2 diabetes; a
1078 randomized, double-blind, placebo-controlled study. Nutr Diabetes. 2018;
1079 8(1):30.
1080 80. Mineo S, Noguchi A, Nagakura Y, Kobori K, Ohta T, Sakaguchi E, et al.
1081 Boysenberry Polyphenols Suppressed Elevation of Plasma Triglyceride
1082 Levels in Rats. J Nutr Sci Vitaminol. 2015; 61:306-312.
1083 81. Nishibori N, Kishibuchi R, Morita K. Suppressive effect of Okara on
1084 intestinal lipid digestion and absorption in mice ingesting high-fat diet. Int J
1085 Food Sci Nutr 2018; 69:690-695.
1086 82. McLaren DG, Han S, Murphy BA, Wilsie L, Stout SJ, Zhou H, et al.
1087 DGAT2 Inhibition Alters Aspects of Triglyceride Metabolism in Rodents but
1088 Not in Non-human Primates. Cell Metab. 2018; 27:1236-1248.e1236.
Page 47 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1089 83. Kuivenhoven JA, Groen AK. Beyond the genetics of HDL: why is HDL
1090 cholesterol inversely related to cardiovascular disease? Handb Exp
1091 Pharmacol. 2015; 224:285-300.
1092 84. Yvan CL, Ranalletta M, Wang N, Han S, Terasaka N, Li R, et al.
1093 Combined deficiency of ABCA1 and ABCG1 promotes foam cell accumulation
1094 and accelerates atherosclerosis in mice. J Clin Invest. 2007; 117:3900-3908.
1095 85. Bonetti PO, Lerman LO, Lerman A. Endothelial dysfunction: a marker of
1096 atherosclerotic risk. Arterioscler Thromb Vasc Biol. 2003; 23:168-175.
1097 86. Ahotupa M. Oxidized lipoprotein lipids and atherosclerosis. Free Radic
1098 Res. 2017; 51:439-447.
1099 87. Greaves DR, Gordon S. The macrophage scavenger receptor at 30 years
1100 of age: current knowledge and future challenges. J Lipid Res. 2009; 50
1101 Suppl:S282-286.
1102 88. Libby P, Ridker PM, Hansson GK. Progress and challenges in translating
1103 the biology of atherosclerosis. Nature. 2011; 473:317-325.
1104 89. Annema W, Von EA, Kovanen PT. HDL and atherothrombotic vascular
1105 disease. Handb Exp Pharmacol. 2015; 224:369-403.
1106 90. Drechsler M, Megens RT, Van ZM, Weber C, Soehnlein O.
1107 Hyperlipidemia-triggered neutrophilia promotes early atherosclerosis.
1108 Circulation. 2010; 122:1837-1845.
1109 91. Serbina NV, Pamer EG. Monocyte emigration from bone marrow during
1110 bacterial infection requires signals mediated by chemokine receptor CCR2.
1111 Nat Immunol. 2006; 7:311-317.
1112 92. Shi C, Jia T, Mendez FS, Hohl TM, Serbina NV, Lipuma L, et al. Bone
1113 marrow mesenchymal stem and progenitor cells induce monocyte emigration
Page 48 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1114 in response to circulating toll-like receptor ligands. Immunity. 2011; 34:590-
1115 601.
1116 93. Brownlee M. Biochemistry and molecular cell biology of diabetic
1117 complications. Nature 2001; 414(6865):813-820.
1118 94. Nakagawa T, Tanabe K, Croker BP, Johnson RJ, Grant MB, Kosugi T, Li
1119 Q. Endothelial dysfunction as a potential contributor in diabetic nephropathy.
1120 Nat Rev Nephrol. 2011; 7(1):36-44.
1121 95. Fraser A, Harris R, Sattar N, Ebrahim S, Davey SG, Lawlor DA. Alanine
1122 aminotransferase, -glutamyltransferase, and incident diabetes: the british
1123 women's heart and health study and meta-analysis. Diabetes care 2009;
1124 32(4):741-750.
1125 96. Chen SC, Tsai SP, Jhao JY, Jiang WK, Tsao CK, Chang LY. Liver fat,
1126 hepatic enzymes, alkaline phosphatase and the risk of incident type 2
1127 diabetes: a prospective study of 132,377 adults. Sci Rep. 2017; 7(1):4649.
1128 97. Kewcharoenwong C, Rinchai D, Nithichanon A, Bancroft GJ, Ato M,
1129 Lertmemongkolchai G. Glibenclamide impairs responses of neutrophils
1130 against Burkholderia pseudomallei by reduction of intracellular glutathione.
1131 Sci Rep. 2016; 6: 34794.
1132
1133
1134
1135
1136
Page 49 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1137 Supplementary Materials:
1138 Supplemental Figure 1: D. indica fruit
1139 Graphical Abstract: Extract of D. indica in water reduces FSG, serum insulin
1140 levels, and ameliorates the serum lipid profile in diabetic model rats without
1141 any adverse effects on kidney and liver tissues.
1142 Extract of D. indica in ethanol significantly reduces serum cholesterol, TG,
1143 LDL with no reduction in HDL levels in hyper-lipidemic model rats.
1144 Highlights
1145 •D. indica fruit extracts diminished fasting serum glucose (FSG) levels in STZ-
1146 induced type 2 diabetic model rats
1147 •D. indica fruit extracts boosted insulin secretion
1148 •D. indica fruit extracts showed no toxic effects on the kidney and the liver
1149 functions
1150 •Extract in water was more effective in reducing FSG levels than extract in
1151 ethanol
1152 •Chronic consumption of 1% cholesterol, 5% coconut oil and 5% cow fat diet
1153 was sufficient to make the rat hypercholesterolemic
1154 •D. indica fruit extract has the potential to reduce serum cholesterol, TG, LDL
1155 with prevention in reduction in serum HDL levels.
1156
Page 50 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
Page 51 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.
1176
1177
Page 52 of 52 bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license. bioRxiv preprint doi: https://doi.org/10.1101/804815; this version posted October 14, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC 105 and is also made available for use under a CC0 license.